Departamento de Estadística
http://hdl.handle.net/10016/12
2014-09-16T04:58:00ZA Bootstrap Likelihood approach to Bayesian Computation
http://hdl.handle.net/10016/19328
A Bootstrap Likelihood approach to Bayesian Computation
Zhu, Weixuan; Marín Diazaraque, Juan Miguel; Leisen, Fabrizio
Universidad Carlos III de Madrid. Departamento de Estadística
Recently, an increasingly amount of literature focused on Bayesian computational
methods to address problems with intractable likelihood. These algorithms are known as
Approximate Bayesian Computational (ABC) methods. One of the problems of these
algorithms is that the performance depends on the tuning of some parameters, such as
the summary statistics, distance and tolerance level.
To bypass this problem, an alternative method based on empirical likelihood was
introduced by Mengersen et al. (2013), which can be easily implemented when a set of
constraints, related with the moments of the distribution, is known.
However, the choice of the constraints is crucial and sometimes challenging in the sense
that it determines the convergence property of the empirical likelihood. To overcome
this problem, we propose an alternative method based on a bootstrap likelihood
approach. The method is easy to implement and in some cases it is faster than the other
approaches. The performance of the algorithm is illustrated with examples in Population
Genetics, Time Series and a recent non-explicit bivariate Beta distribution. Finally, we
test the method on simulated and real data random fields.
2014-09-01T00:00:00ZDisentangled jump-robust realized covariances and correlations with non-synchronous prices
http://hdl.handle.net/10016/19309
Disentangled jump-robust realized covariances and correlations with non-synchronous prices
Vander Elst, Harry; Veredas, David
Universidad Carlos III de Madrid. Departamento de Estadística
We study the class of disentangled realized estimators for the integrated covariance
matrix of Brownian semimartingales with finite activity jumps. These estimators
separate correlations and volatilities. We analyse – in a through Monte Carlo study –
different combinations of quantile-and-median-based realized volatilities, and four
estimators of realized correlations with three synchronization schemes. Their finite
sample properties are studied under four data generating processes and in presence, or
not, of microstructure noise, and under synchronous and asynchronous trading. The
main finding is that pre-averaged disentangled estimators provide a precise,
computationally efficient and easy alternative to measure integrated covariances on
basis of noisy and asynchronous prices. Moreover, the gain is not only statistical but
also financial. A minimum variance portfolio application shows the superiority of the
disentangled realized estimators in terms of numerous performance metrics.
2014-09-08T00:00:00ZA game theoretic approach to group centrality
http://hdl.handle.net/10016/19231
A game theoretic approach to group centrality
Flores Díaz, Ramón Jesús; Molina Ferragut, Elisenda; Tejada, Juan
Universidad Carlos III de Madrid. Departamento de Estadística
This paper is centered in the valuation of the centrality of groups following aproblem-specific approach (Friedkin, 1991). Assuming a TU-game that reflects theinterests which motivate the interactions among individuals in a network, we extend thegame theoretic centrality measure of Gomez et al. (2003) to the case of groups, anddefine the game theoretic group centrality of a group as the variation of its value orpower due to their social relations. We rely on the Shapley group value (Flores et al.,2014) for measuring the value of a group in a game without any restriction, and weintroduce the Myerson group value in order to measure the value when the socialstructure is considered
2014-07-01T00:00:00ZHeterogeneous effects of risk-taking on bank efficiency : a stochastic frontier model with random coefficients
http://hdl.handle.net/10016/19228
Heterogeneous effects of risk-taking on bank efficiency : a stochastic frontier model with random coefficients
Sarmiento, Miguel; Galán, Jorge E.
Universidad Carlos III de Madrid. Departamento de Estadística
We estimate a stochastic frontier model with random inefficiency parameters, which allows us not only to identify the role of bank risk-taking on driving cost and profit inefficiency, but also to recognize heterogeneous effects of risk exposure on banks with different characteristics. We account for an integral group of risk exposure covariates including credit, liquidity, capital and market risk, as well as bank-specific characteristics of size and affiliation. The model is estimated for the Colombian banking sector during the period 2002-2012. Results suggest that risk-taking drives inefficiency and its omission leads to over (under) estimate cost (profit) efficiency. Risk-taking is also found to have different effects on efficiency of banks with different size and affiliation, and those involved in mergers and acquisitions. In particular, greater exposures to credit and market risk are found to be key profit efficiency drivers.Likewise, lower liquidity risk and capital risk lead to higher efficiency in both costs and profits. Large, foreign and merged banks benefit more when assuming credit risk, while small, domestic and non-merged banks institutions take advantage of assuming higher market risk
2014-07-01T00:00:00Z